To support registration and marketing claims, new entrants to the asthma and COPD markets must run comparative trials against market leading products. When aiming to eliminate bias and ensure that the results of these trials are statistically valid, it would be preferable to perform these trials in a double-blind manner. However, how can blinding for products be achieved where physical differences are so extreme? Modification of the inhalation device or using the same device for both treatments is not feasible because the inhaler plays such a critical role in the dispersal and subsequent action of the active ingredient in the airways. This article will look at ways in which blinding of inhalers for double-blind trials can be achieved.
Clinical Study Design
Due to the unique designs of inhalation devices, companies often use a double-blind, double-dummy design to enable them to use their product in its planned market form. This approach also means that they do not need to ‘blind’ their own product as they could produce placebo using the same/similar formulation minus the active ingredient.
While the developer of the innovator product can easily provide an active and placebo of its own inhaler, it is not possible to do this for the comparator. Active comparators may be purchased from the open market; however standard techniques for blinding cannot be used. Placebo units must be produced by converting active inhaler units by removing and replacing their active ingredients and inert materials.
For legal and ethical reasons, it is important that the comparator units used in a clinical trial are distinct from the commercial product. A trial inhaler could contain active product or could be a placebo, so should appear different from the commercial product. This adds work to the production of clinical trial supplies because both the active and placebo product have to be changed in some way.
Blinding Metered Dose Inhalers (MDIs)
These inhalers consist of a plastic body (the actuator) into which an aerosol canister is inserted. Aerosol canisters contain active ingredient and propellants such as Hydrofluoroalkane (HFA). The base of the aerosol may be embossed with a trade name or batch number. Additional markings specific to the product (embossed markings, labels, inkjet printing) may also be present on the actuator body.
Two basic approaches may be used to blinding MDI units. A double-blind design can be used if both the innovator product and the comparator are MDIs with similar size, shape and dimensions. Both products can be encased by a masking device. This is usually a large plastic actuator that covers the entire MDI, while still allowing it to function normally. This approach may be challenging as it’s rare to find two MDIs close enough in size and dimensions. This approach has been used successfully before, but it could be argued that it was too easy for the patient to remove the masking device and potentially unblind the study.
A double-blind, double-dummy design is most commonly used. Contract manufacturers and generics companies involved in MDI manufacture are willing to sell aerosol canisters containing a propellant only. It may be possible to match exactly the size and appearance of the commercial MDI aerosol by identifying the parts used in its manufacture and sourcing identical parts off-the-shelf. Occasionally, the design of specific parts may be covered by a patent or may be produced on a ‘customer-specific’ basis by the manufacturer. If this is the case, parts that are close in appearance but not identical can be used. While the blind may not be perfect, the likelihood of a patient or investigator actually having two different units side-by-side and making a direct comparison should be taken into account.
If such a scenario is unlikely, a slight difference may not be an issue.
If sourcing placebo aerosols from a third party, careful planning is necessary to minimize costs and avoid delays to clinical trials. Set up and changeover costs and times for an aerosol line are significant, so it is important to allow sufficient manufacturing time and effectively plan quantities taking future trial requirements into consideration.
As mentioned above, commercial units may also have text embossed on their base. This can form part of the product identification and therefore should not be copied on placebo units. Placebo units will have blank bases, differentiating them from commercially-sourced actives. To eliminate this difference, it is necessary to cover the bases of both units in the same way. Various methods are available to achieve this.
It is also necessary to de-label commercial units and change the appearance of actuators on active and placebo units so that they are different from those on the commercial product. This generally requires the removal of embossed markings. Since actuator designs are protected, it is not possible to simply copy these. The only way to obtain these for placebo units is to purchase additional commercial units, remove the active aerosol and replace it with a placebo aerosol. One way of limiting this cost is to supply patients with a single actuator and multiple aerosols for a clinical trial.
Dry-Powder Inhalers (DPIs)
Many recently developed inhalers do not use aerosol technology, and involve patients drawing powder into their airways themselves. This approach is believed to improve dispersal of the drug in the airways. Examples of leading DPIs include Advair/Seretide, Pulmicort, Spiriva and Foradil. Each DPI is based on a unique design, so different blinding strategies are required for each.
Active powders for Spiriva and Foradil are provided in hard gelatine capsules. In order to take a dose of these products, patients insert a capsule into the inhaler unit, depress a button to pierce the capsule and inhale the contents through a mouthpiece. Converting the inhaler unit itself for use in a trial is straightforward, and involves removing or obscuring commercial labels or printing on the inhaler body. The difficult part of the process is producing placebo capsules and blisters that match the active commercial product.
Plain capsule shells that match those used in these products can be readily sourced, but lack the printing that appears on the commercial product which cannot be copied. As a result, it is necessary to remove printing from commercial capsules, obscure it or print something similar in appearance onto the placebo capsules. These capsules are typically packaged in high-barrier films, suggesting that removing product from its original packaging may affect its stability. Analytical support for the process is therefore essential.
Capsules should be manufactured using an inert filler that has similar appearance, density and behavior to the active blend contained in commercial capsules. Particle size analysis can help in the selection of a closely matching powder. It is also necessary to choose the correct technology to fill the capsules with an appropriate weight of powder. Fill weights for these products are generally very low, and there are several options now available to manufacture large volumes of capsules with accurate fill weights using modern modular based capsule-filling equipment or smaller volumes for early phase studies on equipment such as the Xcelodose system.
Packaging the placebo capsules in a manner that matches that of the active commercial product can be challenging, as branded blister packs cannot be copied. It is also possible that blister-packaging technology not available to most clinical supplies departments was used to produce the blisters (such as tropicalized blister packs). Again, this can make it difficult to produce matching packs. Therefore, the options are to remove or obscure branding from commercial packs, while ensuring that the approach used does not affect product integrity. Repacking the active and packing the placebo in plain packaging is also an option, however there may be stability issues with such an approach, so analytical support for repackaging is essential.
Some DPIs, such as AstraZeneca’s Turbohaler design or the Asmanex Twisthaler, contain a reservoir of powder containing the active ingredient. In principle, it should be possible to convert these units to placebo simply by disassembling them, removing the active powder and then reclosing the units. However, there are complications in this process:
- Opening the units without causing damage requires special tooling
- Wash methods need to be developed to remove all traces of the active ingredient from the inhaler unit; rinse methods must be developed to ensure that solvents are also removed.
- Containment/extraction units are required to remove active dust and solvent from the atmosphere during the process.
- Analytical tests required to prove that active ingredient has been removed.
- Specialized equipment and tooling is required to re-assemble the units, again without damaging them.
Typically the active powder in the reservoir is replaced with an inert placebo powder, but this approach is not always used. Powder doses delivered by this device are so low that they are believed to be undetectable to the patient. As with other inhalers, commercial labels and markings need to be removed or obscured on each unit so that clinical trial units appear different from commercial units.
In the therapeutic area of asthma and COPD, the best known inhaler unit employing blister strips to deliver active powder for each dose is GSK’s accuhaler/diskhaler. This unit employs a 60-dose blister strip, with a specific dose of drug in each blister pocket. Each time a patient takes a dose he pulls a trigger, which advances the blister strip to the next full pocket and the dose counter to the next number. Disassembly and reassembly of this DPI is highly complex, and therefore the most difficult DPI to convert to placebo. GSK has invested millions of dollars in a robotic line for filling blister strips and assembling these units. Replicating this technology in a typical clinical supplies unit for a few hundred, or indeed thousand, placebo units for clinical trials is not feasible. While alternative approaches can be developed, these still involve significant time, cost and complexity that may exceed the resources and funds available to an in-house clinical supplies unit.
Use of a double-blind, double-dummy design will enable companies developing new inhalation products to perform unbiased comparative trials against market-leading products. However, significant engineering and analytical work is required to convert commercial inhalers to placebos. As a result, it is essential that planning for this type of project begins well in advance of any proposed clinical trial start date. Even if time is available, the cost of automating such a process for a one-off comparative study is huge. If a trial requires only a few thousand comparator inhaler units to be converted to placebo, this can result in very high costs for each unit produced. It is worth bearing in mind that while this cost may be a barrier to an in-house clinical supplies department working on a limited number of trials, it may be less of a barrier to a contractor working with several companies with several trials each. By spreading the cost of developing and equipping this process across several clients and trials, a contractor may be more willing to invest in developing one of these conversion processes. From the industry’s perspective, outsourcing production of placebo units could bring double-blind inhaler studies within reach.
- Inhalation & Nasal Spray Generic Drugs 2012, Drugs, 2010. http://www.espicom.com/prodcat2.nsf/Product_ID_Lookup/00001693?OpenDocument
Martin Lamb is vice president Business Development, at Almac. He can be reached at email@example.com.